There are limited options available to detect impairments impacting driving when impairment is due solely to cannabis use or the use of other psychoactive drugs. In a 2007 roadside survey more drivers were positive for drugs than alcohol (NHTS 2007) and this contributes to the risk of accidents when driving. Cannabis users suffer auto injury ten times more frequently than non-users. (ScienceDaily) Eleven percent of those presenting in an emergency room after an auto accident were found to have used cannabis in the absence of other drugs or alcohol and acute cannabis consumption presents a risk for motor vehicle crash and fatal collisions involving motor vehicles. (Asbridge 2013)
A survey of drivers in Canada found that 4.8 percent of drivers admitted to driving within two hours of using cannabis. (Porath 2013) With the current legalization of cannabis for recreational use in several states, the rate of automobile injury secondary to cannabis intoxication is likely to increase.
The state of Colorado has the most experience within the United States, with cannabis use and driving. Starting Jan. 1, 2014 Amendment 64 took effect allowing for the legal use of recreational cannabis in the state of Colorado for citizens over the age of twenty one. Even though the laws in Colorado clearly state that it is illegal to drive high, twelve percent of the citations issued in 2014 for DUI involved cannabis use. The DUI fatality rate increased by 6 percent in 2014, and the rate of DUI citations increased by 23 percent compared to 2013. Of the total citations issued, over six percent were suspected to involve only cannabis and no other substances. Colorado officials take the position that any amount of cannabis can impair a person for driving and DUI citations can be based on observations. Colorado has established a legal level of cannabis that is considered to impair driving function at five nanograms of active tetrahydrocannabinol, the component impacting cognitive function, per milliliter of whole blood. (Blood) Other states have adopted a zero tolerance level and there is research supporting this that indicates how little cannabis is required to significantly impair skills required for driving. (Impair) The Colorado Department of Transportation did a survey on the attitudes and behaviors of residents several months before recreational cannabis became legal. They found that twenty percent of the respondents that had used cannabis in the previous year had driven shortly after consuming cannabis and those that did so within two hours drove on average seventeen times during a month. (CDOTNEWS) Colorado invested one million dollars on a public education campaign, “Drive High, Get DUI”, in an effort to minimize the impact of driving while impaired secondary to cannabis. (Education) Even after the campaign a survey found that 57% of those reporting to use cannabis, drove within two hours of consumption. (CDOT57)
Washington State has had legal marijuana for a shorter period, but still is reporting significant problems with increasing numbers of impaired drivers due to cannabis consumption. A report (Survey) for the Washington Traffic Safety Commission in a survey conducted by the Pacific Institute for Research and Evaluation found that seventy percent of drivers questioned had used cannabis and of those reporting use 44% acknowledged they had driven a car less than two hours after using cannabis. Further 90% of those who drove while using cannabis, did not feel that cannabis impaired their ability to drive. In the year 2012 there were 988 driver tests that came back positive for cannabis and this increased to 1, 362 in 2013. (KXLY) From 2009 through 2013, more than 1,000 people died in impaired driving collisions in Washington. Impaired driving is involved in nearly half of all traffic deaths and more than 20 percent of serious injury collisions. (Kirkland) When cannabis is combined with alcohol the effect is more deadly than alcohol alone. Alcohol intoxication increases the risk of a fatal accident by thirteen times compared to a sober driver but alcohol and cannabis intoxication increases the risk twenty four times. (Brady)
The National Institute on Drug Abuse prepared a White Paper on drugged driving research in 2011 that discussed marijuana use; “For illegal drugs, Zero Tolerance (ZT) per se laws are those which set that limit at the drug detection cut-off level. In concept it is not necessary to prove driver impairment to convict an offender under a per se law.” (NIDA/Whitehouse} They further discussed per se drugged driving laws, “A per se drugged driving law is one in which a specified level of a drug in the body of a driver is defined as an offense. This may be a level at which here is evidence that the drug has been shown to effect driver performance such as the 0.08 g/mL limit for alcohol.”
There is limited research on cannabis impairment related to driving. Impairment due to cannabis differs significantly than the impairments caused by alcohol. Even small amounts detected in the blood have been shown to impair the cognitive perceptual functions necessary for safe driving. (Zero). The demonstration of cognitive dysfunction even with small amounts of cannabis is fueling arguments for zero tolerance of cannabis with driving. Fifteen states have zero tolerance regulations when it applies to drug use and cannabis. (Fifteen) While some exclude medical marijuana users, an estimated nine have more absolute laws with zero tolerance applied to all cannabis users.
In a September 2015 report from the Governor's Highway Safety Administration the complexities of evaluating impairment and driving under the influence of cannabis were discussed. Extraordinary attention was paid to cannabis as it is considered to be a threat to public health safety. The report reiterates that driving under the influence of drugs; DUID, is illegal in every state. This is much the same as driving under the influence of alcohol, DUI. Like driving under the influence of alcohol, DUID has two requirements that law enforcement must take into consideration. The driver must exhibit signs of impairment through behavior observed by a law enforcement officer and the impairment must be linked to a drug. An officer must have observed a driver demonstrating impairment. Only then can an officer obtain chemical evidence of a drug, usually through a blood test, and the officer must be able to link drug presence to the observed impairment. If the driver refuses a blood test, the officer relies on observations. All this takes longer than for alcohol. With alcohol use the signs are well understood and backed by years of research. The Standardized Field Sobriety Testing (SFST) is an efficient and accurate screening. Evidence of blood alcohol level can be obtained by the biomarkers present in a breath test. The links between SFST, breath-testing and alcohol impairment are recognized by the judiciary system. Traditional training of law enforcement officers does not always include adequate training to observe impairment to driving that arises from drugs. It can take several hours to obtain the needed legal permissions to draw blood from a driver and there are concerns that blood levels related to cannabis do not always match the impairments. Further drug testing can be expensive with labs having substantial backlogs. Cannabis related driving impairment presents a unique challenge to law enforcement as well as to the court systems. A clear understanding of how cannabis has an impact on the ability to drive and limits on how an officer can compel further testing creates barriers to removing drivers impaired by cannabis consumption from the road. Unless there is evidence of impairment to drive similar to the SFST related to alcohol, an officer cannot require a driver to undergo further evaluations.
When there is a concern related to impairment to drive, police officers use standardized observation techniques. The most common is the SFST. The assessments used in the SFST were developed primarily to determine alcohol related impairment to drive. Currently assessments of the visual system by observation are part of a standardized field sobriety test sequence used to detect alcohol impairment. The standardized field sobriety test sequence is used to detect other impairments; including cannabis.
A part of the SFST is the screening for alcohol related horizontal gaze nystagmus. The horizontal gaze nystagmus related to alcohol occurs when the eyes move from looking straight ahead to the side in a horizontal motion. The observation of a driver's eye movements along the horizontal axis while the driver is fixating a target are subjectively interpreted for the presence of horizontal nystagmus, jerking eye movement, and eye pursuit dysfunctions. If the nystagmus is observed to be present when the eye position is at an angle approximating forty five degrees or less and/or there are losses of fixation during pursuit eye movements there is a high likelihood that the driver has a blood alcohol level that would impair driving.
Horizontal gaze nystagmus testing has been shown to have an accuracy of over 75% in the detection of blood alcohol content of 0.09% or greater in a study undertaken by Dixon and colleagues. (Dixon 2009) Furthermore, studies of horizontal gaze nystagmus have shown the test to be sensitive even if blood alcohol levels are lower than legal limits of 0.08%. Horizontal gaze nystagmus testing, when properly administered is sensitive in blood alcohol levels of 0.04-0.08%. (McKnight 2002) The horizontal nystagmus test is not complex to administer, but it does require significant learned skill to interpret a driver's response. This can create issues when presented in court as the officer's administration and interpretations of the test are subjective. When the signs of horizontal gaze nystagmus occur, there is concern that the blood level is greater than 0.04%. (McKnight 2002) The rate of alcohol related auto accidents is staggering. Over 15,000 deaths associated with alcohol related traffic accidents occur annually, and over 40% of auto fatalities each year are related to alcohol consumption. Cannabis also impairs driving but, unlike alcohol, it is difficult to determine blood levels that cause impairment.
Application of eye movement as an indicator of impairment for cannabis use is inconclusive. Citek and colleagues evaluated 25 participants identified by urinalysis to have cannabis as the only intoxicant and found that there was lack of findings for deficits in eye movements. (Citek 2012) Adams and colleagues found deficits in horizontal gaze nystagmus (HGN) as well as visual tracking with the consumption of cannabis, but not at a level as significant as those seen with intoxication with alcohol. (Adams 1975) Smooth tracking eye movements and saccadic tracking eye movements are reduced with alcohol but not with marijuana or a placebo in the motion study undertaken by Flom and colleagues. (Flom 1976) In a study of 20 adults using cannabis and cannabis combined with alcohol, researchers found that with cannabis alone the users showed impairment with the field sobriety test of one leg stand, but dysfunctions in regards to horizontal gaze nystagmus were when cannabis was combined with alcohol. (Bosker 2012) Accordingly, horizontal gaze nystagmus appears not to be affected by cannabis even though affected by alcohol. One leg standing is uncertain as to its correlation with cannabis consumption.
There are products on the market to test for drug use. One the DrugTrap®, consists of a filter holder, mouthpiece, plastic bag with volume indicators and seals for both ends. The consumer blows into the mouthpiece and then the bag is sealed and mailed to the company for analysis. The product is currently not FDA approved. (DrugTrap) Another product, though not on the market, is a telephone application to record and analyze eye movement scanning with cannabis consumption. This is described by Arizona State University Center for Innovation. (Arizona) A further product, BreathalEyes records eye movements and nystagmus and calculates the probable blood alcohol content of the person using the application on a smart phone or tablet. (BreathalEyes) MyCanary is yet another product. MyCanary is a telephone application that allows a driver to test their own theoretical potential to be too impaired to drive. It utilizes simple cognitive and physical tests including balance, memory, reaction and time perception and provides a readout for the driver. There is no research or science background available on this product and it is not intended to be used by law enforcement or the justice system. (MyCanary) Canary)
Background on Retina and Brain Nuclei related to Cannabinoid Receptors
The active ingredients in cannabis act on cannabinoid receptors in the human body. The primary cannabinoid receptors that have been identified are classified as either CB1 or CB2 receptors. The CB2 category is highly represented in the central nervous system; including the retina. CB2 receptors are represented in the central nervous system but to a lesser degree than CB1 receptors. CB1 receptors have their greatest prevalence in the periphery and immune systems. In humans there are two primary endogenous compounds acting on the receptors, N-arachidonoylethanolamine (anandamide, AEA) and 2-arachidonoylglycerol (2-AG). Anandamide acts primarily post-synaptically as a retrograde compound to modulate neurotransmitters. Anandamide has greater affinity for CB1 receptors. 2-AG has been found to act pre-synaptically and also has greater affinity for CB1 receptors. 2-AG is found abundantly in the brain, but shows less affinity for CB1 receptors than does Anandamide. (Shwitzer 2015) The receptor sites in the brain related to CB1 and CB2 are primarily those involved in higher cognitive functions. The forebrain, midbrain and hindbrain have areas associated with the control of movement that are affected and hindbrain areas associated with the control of motor and sensory functions of the autonomic nervous system are affected. (Glass 1997) All regions where cannabinoid receptors have been identified have implications for performance related to driving.
The human retina has representation of cannabinoid receptors throughout multiple layers and cell structures. This is supported by animal models. CB1 activity in human retina is evidenced by staining in the synaptic layers of the retina; the inner and outer plexiform layers. The density of CB1 Receptors increases in the inner nuclear layer and the ganglion cell layer. There is substantial staining in the outer segments of the photoreceptors. (Straiker 1999) Research has demonstrated the expression and regulation of CB1 receptors in human retinal pigment epithelium cells. (Wei 2013) An animal model using mice supports this finding with identification of CB1 receptors in the inner retina and ganglion cells, with integrations and processing of excitatory signal from bipolar cells and inhibitory signals from amacrine cells. (Wang 2013) The activation of CB1 receptors differs dependent on the circadian quality of light; night versus daytime. If CB1 receptors are activated during day, the rod-cone gap junctional signaling is decreased. However if activated at night the rod-cone gap junctional signaling is increased. (Jieng ARVO) This has functional implications for scotopic vision and glare recovery.
An additional rodent model has demonstrated that CB2 receptors are localized in cone and rod photoreceptors, horizontal cells, some amacrine cells, and bipolar and ganglion cells. (Cecyre 2013) Additional evidence of CB2 receptors within rodent retina as well as the central nervous system has been identified by additional groups. (Hu 2010) Lu and colleagues identified CB2 evidence in the somas of retina ganglion cells in a rodent model. (Lu 2000) This differs from a primate model which shows that CB2 receptors are in the primate retina but exclusively in the retinal glia, with the model still supporting that CB1 receptors are present in neuroretina. (Bouskilla 2013CB2)
One of the primary brain nuclei involved in processing visual signals is the lateral geniculate nucleus (LGN) and this area of the brain is dense in cannabinoid receptors. A primate model; the vervet monkey, shows that CB1 receptors are located throughout the LGN with prominent findings in the magnocellular layers. The receptors are less prominent, but still evident, in the koniocellular layers. (Javadi 2015) Magnocellular functions involve primarily achromatic signals; related to contrast and temporal functioning of vision. The currently used testing for glaucoma utilizing contrast and temporal functioning is assessing magnocellular processing. Koniocellular functioning involves the processing of chromatic signals in the blue wavelength. Another primate model of CB1 functioning within the LGN shows demonstrated that the active ingredient in cannabis inhibits cells that would normally fire when exposed to light and cells that would be inhibited by light were either unresponsive with no inhibition activity or actually had an increase in excitation. (Bieger 1972) This has significant implications for the interaction of central macular retinal functions and peripheral retinal functions as well as scotopic (night vision or dim light vision) and photopic (daylight vision or bright light vision) functions.
DaSilva and colleagues were able to quantify the functional action on CB1 receptors within the LGN and found two populations; 28% were excited by an antagonist and 72% were inhibited. When activated artificially (as they would be with cannabis) the visual signals were altered. With excitatory activity there was a decrease in the signal to noise ratio but an increase in variability. With altered inhibition; which accounts for over seventy percent of the cells in the LGN, there was an increase in the signal to noise ratio with reductions in variability. The researchers concluded that the abnormal signals originating from the LGN with artificial stimulation of the cannabinoid receptors using cannabis and then traveling to the cortex would account for the behavioral effects of cannabis. (DaSilva 2012) The findings in the LGN support evidence of diverse roles of cannabinoid receptors in both the retina and the LGN, in modulating both excitation of cells and the inhibitory cell functions. The authors hypothesize that the cannabinoid receptor functions within the visual system account for many of the behavioral effects from cannabis. The behavioral effects and changes in cognitive function along visual pathways, demonstrated by functional brain imaging, are enough to impair driving functions. (MRI) A rodent model of development and function of CB1 receptors in the visual cortex found intense staining for CB1 receptors in layers II, III, and VI. The functions were influenced by dark and light cycling and had plasticity related to retinal stimulation. (Yoneda 2013)
Reductions in acuity have been reported secondary to cannabis consumption. (Dawson 1977) Adams and colleagues did not find reductions in static acuity but did find reductions in dynamic acuity after cannabis consumption. (Adams 1975)
The structural findings related to the cones in the retina and koniocellular layers of the LGN offer an explanation for the functional findings of color impairment with cannabis consumption. Several researchers have identified color deficits along the blue axis. Adams and colleagues found dose related impairment with the consumption of cannabis was identified using the Farnsworth-Munsell 100 hue test and the findings were along the blue axis. The deficits were similar to those blue deficits that occur with retinal based pathology leading researchers to conclude that the origin of the dysfunction was in the retina itself (Adams 1976) Dawson and colleagues supported the findings of significantly reduced color vision functions with decreased color matches among those having consumed cannabis. (Dawson 1977) In studies of retinal tissue in vitro; Hu and colleagues found evidence of cannabinoid function in postsynaptic cone bipolar cells that interact with cone photoreceptors providing further physiologic evidence to support the functional deficits in color processing. (Hu 2010)
Early reports in the 1970s indicated that marijuana impacted pupillary function with dose related constrictions of the pupil. (Hepler 1972, Brown 1977) More recent reports using pupilometer technology are documenting dose related dilations of the pupil. (Stark 2003, Merzouki 2008) Dilated pupils as well a slow pupil reaction were reported to be indicators of cannabis consumption by Bramness and colleagues. The diminished pupil reaction persisted for the first two hours. They observed an increase in dilation among those with blood cannabis concentrations above 2.9 ng/ml, but the observation was only present in 35% of those consuming cannabis. (Bramness 2010)
There is evidence of dysfunction related to dark adaptation, light adaptation, glare recovery and photopic functions with the consumption of cannabis. (Dawson 1977, Adams 1978) This may be related to changes in pupil function, suppression of central inhibitory or excitatory retinal functions, abnormal retinal functions peripherally or any combination of the foregoing. The dysfunctions persist for two hours after cannabis consumption. That cannabis impacts central photopic based functions is evidenced by studies that demonstrate increase in scotopic functions. There are reports of improved night vision with the use of cannabis. (Russo 2004) In support of increased peripheral scotopic function is the discovery of a novel exogenous cannabinoid in rod segments and elsewhere in the central nervous system of a primate model, GPR55. (Bouskilla 2013GPR55)
There is room for improvement in the area of testing for functional impairment due to the consumption of cannabis and other psychoactive drugs. There is currently no field applicable test available to identify those driving impaired due cannabis consumption.